Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/rahmanef63/control-room/claude-mdgit clone --depth 1 https://github.com/rahmanef63/control-roomWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/rahmanef63/control-room/claude-md)<a href="https://agentmods.dev/instructions/rahmanef63/control-room/claude-md"><img src="https://agentmods.dev/badge/instructions/rahmanef63/control-room/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00997 | $0.00997 |
| Opus 5 | $0.00498 | $0.00498 |
| Sonnet 5 | $0.00199 | $0.00199 |
| Haiku 4.5 | $0.00100 | $0.00100 |
Grade A, and why
control-room CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Control Room — Canonical Agent Instructions
Current baseline: v2.0.0, terminal-first.
Product boundary
Control Room is a tmux-like terminal multiplexer with a browser/PWA UI.
Core product scope:
- PTY lifecycle and reconnect;
- multiple terminal panes;
- workspaces;
- single/grid/fullscreen terminal views;
- mobile terminal UX and safe areas;
- rename/duplicate/move/focus/close;
- terminal history and restore;
- terminal templates;
- cwd/file-picker helpers;
- broadcast input;
- lightweight terminal/host status;
- optional CLI launch profiles and activity decoration.
Out of scope:
- provider/account or credential management;
- browser automation;
- cron/job orchestration;
- AI supervisory orchestration;
- managed application lifecycle platforms;
- project-specific runtime dependencies;
- deployment orchestration unrelated to deploying Control Room itself.
A CLI inside a pane manages its own auth, tools, browser work, deployments, and agent behavior.
Architecture
browser / PWA
-> SvelteKit 2 + Svelte 5 frontend (Node 22 production runtime)
-> authenticated HTTP proxies
-> terminal output to browser over SSE
-> frontend server is the WebSocket client to the agent
-> Node 22 host agent on loopback
-> node-pty
-> bounded host/file/state helpers required by terminal UX
The frontend never shells out directly and never exposes AGENT_GATEWAY_SECRET or CONTROL_ROOM_SECRET to browser JavaScript.
Standalone rule
The repository must remain independently understandable and runnable. Do not add architecture, docs, configuration, imports, or runtime requirements that assume another internal repository exists.
Optional third-party CLI programs may be launched as normal terminal processes. Their existence is not a Control Room runtime dependency.
Frontend rules
- SvelteKit 2 + Svelte 5 runes.
- Use
$props(),$state,$derived,$effect, snippets, and rune-backed state modules. - Do not introduce legacy
export let,$:,on:click, legacy<slot>, or writable/readable store patterns in modified code. - UI primitives:
frontend/src/lib/components/ui/. - Terminal features:
frontend/src/lib/features/terminals/. - Server-only helpers:
frontend/src/lib/server/. - Route APIs:
frontend/src/routes/api/**/+server.ts. - Preserve safe-area, mobile keyboard, fullscreen, viewport and xterm-fit behavior.
- Preserve the app-chrome hierarchy: compact app/status bar → workspace rail → session/command bar → terminal stage; collapse labels/secondary utilities before allowing toolbar wrapping.
- Keep
/landingpresentation-only and public: prerendered,csr = false, no host/gateway reads, no external runtime resources, route-specific SEO/structured data, andprefers-reduced-motionsupport. - Delete obsolete implementations; do not keep compatibility placeholder files.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed · -13 lines · -1,081 tokens per session scan B → A c0ed1201f366
- 5d ago First seen · 136 lines · 2,078 tokens per session scan B 7dec43cb188a
control-room CLAUDE.md is an instructions file published in the GitHub repository rahmanef63/control-room (19 stars, last pushed yesterday), licensed MIT. It adds 997 tokens to every session, about $0.0050 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
serverless-full-stack-webapp-starter-kit AGENTS.md
AGENTS.md instructions for aws-samples/serverless-full-stack-webapp-starter-kit, covering agents.md, commands, install all dependencies, webapp and cdk.
mastra AGENTS.md
AGENTS.md instructions for mastra-ai/mastra: Unless asked, don't inspect reference or modify examples. Use the most-specific AGENTS.md; for package work, read packages/ /AGENTS.md first.
souls-directory CLAUDE.md
Instructions for thedaviddias/souls-directory, covering claude.md — project context for ai assistants, project overview, tech stack, critical rules and database is convex.
agentgram CLAUDE.md
Claude Code instructions for agentgram/agentgram, covering claude.md — agentgram development guide, project overview, tech stack, monorepo structure and core architecture patterns.
nextarter-tailwind AGENTS.md
AGENTS.md instructions for agustinusnathaniel/nextarter-tailwind, covering next.js: always read docs before coding, ai agent guidance: nextarter-tailwind, 1. mental model for agents, implementation delegation and tooling strategy.
Silex AGENTS.md
AGENTS.md instructions for silexlabs/Silex, covering agents.md, tech stack, run it locally, building the desktop app and when writing code for silex (editing source).